Czar Not Found about arvo

[jfranklin9000]

The time between retries increases.

[joemfb]

This normally just means domain-name resolution for these galaxies failed. Once you try to send a packet to a ship, %ames (the urbit network) will keep trying indefinitely (subject to backoff).

But I've never seen it repeat endlessly. Have you tried to send lots of packets to a variety of ships?

[minderfolden]

Not to the extent that would make this happen. This happens all day long. I fire up the screen and I see hundreds of lines of it.

[vvisigoth]

Something weird here with the back off.

On Wed, Jul 3, 2019 at 6:59 PM custom-jonathan ***@***.***> wrote: [image: unknown] <https://user-images.githubusercontent.com/51895363/60634673-620b1880-9dd5-11e9-9bf2-578bbe620b1d.png> Not to the extent that would make this happen. This happens all day long. I fire up the screen and I see hundreds of lines of it. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1211?email_source=notifications&email_token=AAMJBY6U5A6BJLKS2EASIHTP5VKQDA5CNFSM4H5LX2Q2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZGDEWI#issuecomment-508310105>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAMJBY3YZAACP7Z6UJBHVETP5VKQDANCNFSM4H5LX2QQ> .

-- — ~poldec-tonteg http://urbit.org

[minderfolden]

I'm not familiar with the "with the back off" saying. What do you mean?

[vvisigoth]

Yes. If a request fails then the retries should decrease in frequency each time. But that appears to not be happening.

On Thu, Jul 4, 2019 at 9:26 AM jfranklin9000 ***@***.***> wrote: The time between retries increases. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1211?email_source=notifications&email_token=AAMJBY6TOVXAYZALIT7ERYTP5YQE5A5CNFSM4H5LX2Q2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZHZURI#issuecomment-508533317>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAMJBY65LGZKLM4RQE6CNTTP5YQE5ANCNFSM4H5LX2QQ> .

-- — ~poldec-tonteg http://urbit.org

[vvisigoth]

(mirrored from email)

I have your pier and ran it. Nothing seems amiss, no recurring errors, etc. why is this?

My guess (and some other engineers) is that vere (the runtime) has its own codepath for galaxy discovery and, thus, its own way of dealing with these requests. I'd guess that this is a deep bug in the vere runtime, not in your pier. This is good news and means that me using 1) a different ip and 2) a different instance of vere with its own cache, seemed to help.

[vvisigoth]

I'm going to close this since we have a preliminary work around and we're changing this codepath.